Evaluation method of craniofacial asymmetry index based on artificial intelligence
Abstract
An evaluation method of craniofacial asymmetry index based on artificial intelligence is disclosed and includes: a craniofacial image shooting step: obtaining a craniofacial model file of a patient; an artificial intelligence head shape identification and feature point marking step: importing the craniofacial model file into an artificial intelligence algorithm, performing identification and feature point marking on a craniofacial image in the craniofacial model file to generate at least one feature point; a craniofacial space coordinate axis establishment step: including a coordinate axis y-z plane establishment step, a coordinate axis origin establishment step and a z-axis orientation definition step; and an artificial intelligence skew degree estimation step: inputting the craniofacial model file and corresponding coordinate axes into an artificial intelligence skew degree evaluation algorithm simultaneously, and presenting a craniofacial skew degree in a data visualization manner.
Claims
exact text as granted — not AI-modifiedWhat is claimed is:
1 . An evaluation method of craniofacial asymmetry index based on artificial intelligence, comprising:
a craniofacial image shooting step: obtaining a craniofacial model file of a patient; an artificial intelligence head shape identification and feature point marking step: importing the craniofacial model file into an artificial intelligence algorithm, performing identification and feature point marking on a craniofacial image in the craniofacial model file to generate at least one feature point; a craniofacial space coordinate axis establishment step: including a coordinate axis y-z plane establishment step, a coordinate axis origin establishment step and a z-axis orientation definition step; and an artificial intelligence skew degree estimation step: inputting the craniofacial model file and corresponding coordinate axes into an artificial intelligence skew degree evaluation algorithm simultaneously, and presenting a craniofacial skew degree in a data visualization manner; wherein the coordinate axis y-z plane establishment step is to select a brow center, a most concave point of a bridge of a nose and a tip of the nose to define a first symmetrical plane as a coordinate axis y-z plane; wherein the coordinate axis origin establishment step is to select left and right ear holes on two sides, project the left ear hole and the right ear hole respectively to the coordinate axis y-z plane to obtain two projected points, take a midpoint position of the two projected points and define the midpoint position as a coordinate axis origin; and wherein the z-axis orientation definition step is to identify a line formed by the most concave point of the bridge of the nose and the coordinate axis origin, and define the line as a z-axis orientation.
2 . The method according to claim 1 , wherein the craniofacial image in the craniofacial model file includes a face, a head and a back of a skull.
3 . The method according to claim 1 , wherein the craniofacial image shooting step is performed by a three-dimensional photography device.
4 . The method according to claim 1 , wherein in the artificial intelligence head shape identification and feature point marking step, the artificial intelligence algorithm first identifies a skew type of the craniofacial image from the craniofacial model file, and then automatically identifies the skew type of the craniofacial image, and the at least one feature point is marked as a basis for the craniofacial space coordinate axis establishment step.
5 . The method according to claim 1 , wherein the artificial intelligence skew degree evaluation algorithm in the artificial intelligence skew degree estimation step calculates an asymmetry value of a craniofacial size based on the inputted craniofacial model file and a skew type and an overall space vector and presents a degree of craniofacial skew in a data visualization format.Cited by (0)
No later patents cite this yet.
References (0)
No backward citations on record.